MR Scanner Single Point Imaging K-Space Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Single point imaging in magnetic resonance (MR) is excessively time-consuming, requiring 30 minutes to acquire MR data at an isotropic resolution of 1 mm and a matrix size of 128*128*16, making it impractical for clinical use.
Innovation Solution
The method involves acquiring MR data by radiating an RF excitation pulse and applying magnetic field gradients for spatial encoding in only two directions, allowing for the acquisition of a single k-space point, significantly reducing scan time by limiting data acquisition to a plane rather than three-dimensional space, and combining this with line-by-line acquisition outside the central k-space region.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If single point imaging is used to maintain constant encoding time for all k-space points, then robustness against magnetic field inhomogeneities is improved, but scan time increases excessively
Solution Approach 1:
The patent segments k-space into a central region and peripheral regions. The central region is acquired using single point imaging with constant encoding time to maintain robustness, while peripheral regions use accelerated acquisition methods. This segmentation allows different acquisition strategies to be applied to different parts of k-space, balancing robustness and scan time.
Solution Approach 2:
The patent applies single point imaging selectively to only the central region of k-space rather than the entire k-space. This partial application of the robust acquisition method reduces the total number of acquisitions required while maintaining the benefits of constant encoding time where they are most needed (at the center), thereby reducing overall scan time.
2Loss of information
If k-space points are acquired in three-dimensional space, then complete volumetric information is obtained, but data acquisition time increases
Solution Approach 1:
The patent reduces the acquisition from three-dimensional k-space sampling to two-dimensional plane sampling by exploiting the slice selection capability of MRI. By acquiring data in a 2D plane and using slice selection gradients, the method obtains volumetric information through a lower-dimensional acquisition approach, significantly reducing scan time while maintaining essential volumetric imaging capability.
3Measurement precision
If magnetic field gradients are applied for spatial encoding in three directions, then complete spatial resolution is achieved, but encoding complexity and scan time increase
Solution Approach 1:
The patent segments the spatial encoding into two stages: first, 2D spatial encoding in the imaging plane using magnetic field gradients in two directions; second, slice selection encoding in the third direction using a slice selection gradient. This segmentation simplifies the simultaneous 3D encoding problem into manageable 2D plane encoding plus slice selection, reducing gradient switching complexity.
Solution Approach 2:
The patent handles the third spatial dimension through slice selection rather than through gradient-encoded spatial encoding. By using a slice selection gradient during RF excitation to define the imaging plane, the method reduces the encoding complexity from three gradient-encoded dimensions to two gradient-encoded dimensions plus one selection dimension, simplifying the overall encoding scheme.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach significantly reduces MR data acquisition time, avoids fold-overs in the third spatial direction, and minimizes the impact of magnetic field inhomogeneities and chemical shift artifacts, enabling faster and more efficient MR imaging.
Implementation Method 1
An RF excitation pulse is radiated into a subject. Apply magnetic field gradients for spatial encoding of the magnetization of nuclear spins in the subject excited by the RF excitation pulse
Implementation Method 2
Apply magnetic field gradients for spatial encoding of the magnetization of nuclear spins in the subject excited by the RF excitation pulse, wherein only two directions in space (and not the third) are encoded by this spatial encoding
Data Source
AI summary
In a method and a magnetic resonance (MR) scanner for producing an MR image of a volume portion of a scanned object, MR data are acquired by repeatedly executing the following step of radiating an RF excitation pulse, applying magnetic field gradients for spatial encoding of a magnetization excited by the RF excitation pulse, with only two directions in space being encoded by the spatial encoding, and acquiring essentially a single k-space point for each radiated RF excitation pulse. An MR image is reconstructed from the MR data thusly acquired.